did. We were just hours away from closing every stock and buy market in the world. Can you explain to our listeners a little bit, so for people to understand that, and two points, I actually want to say something. LTCM was using a Gaussian distribution in their model for trading, which is an important point that will circle back around when we speak about predictive modeling. Gaussian distributions are normal distributions. The idea that if you have a certain number of events that are out of the norm, you can keep betting that you're going to get back into the average. And they were doubling down, tripling down that things were going to recur. It's a bell curve. It's mean reverting. It's called a Gaussian distribution, or a normal distribution. It's got a couple different names, but that's exactly right. And the problem with that, first of all, it's not true. It's empirically not true. We don't have to debate it. If you look at it. It's true in the case of people's height, in the case (17/98)
You are viewing a single comment's thread from: